Venom immunotherapy reduces the risk of a systemic allergic reaction to an insect sting from the 50% level observed before treatment to less than 2% on maintenance therapy. When therapy is discontinued after 5 years or more, we have previously reported that the risk of a sting reaction remains less than 10% in the first few years after stopping treatment. Some investigators have studied discontinuation of therapy after only 3 years, but have suggested restrictions such as younger patients, milder reactions, negative sting challenge during therapy or negative RAST. In the first 3 years off therapy, our results showed that the venom skin test and IgE antibody levels became progressively weaker or negative, and the patients who did have systemic reactions to challenge stings had mild or even trivial symptoms. Most of the studies of discontinuing venom immunotherapy have reported the outcome of stings only in the first year or two after stopping. There are no reports of the long-term outcome in patients who remain off treatment. We hypothesized that there would become almost no risk of reaction with the disappearance of sensitivity over time in patients who stop treatment after 5 years. However, recent reports suggest that the risk may persisit and that the severity of the pre-treatment reaction may be a determinant of the risk of reaction. It has been reasonably assumed that treatment can be stopped if the venom skin test or RAST becomes negative, which occurs in a minority of patients. However, the level of skin test sensitivity to Hymenoptera venom, the best available indicator, does not correlate with the severity of the clinical reaction to a sting and no markers have been found to predict the outcome of a sting any more accurately than the average 50% risk predicted by history and skin test. The results of this study will show whether patients can safely stop after only 3 years instead of our current recommendation of 5-6 years, and whether the risk of reaction remains acceptably low in the 10 years after stopping therapy. We will also determine whether there is a lower risk in those patients with a less severe history of reaction before treatment or in those who have developed negative skin tests and/or RAST.